Summary of Mitigating Catastrophic Forgetting in Language Transfer Via Model Merging, by Anton Alexandrov et al.
Mitigating Catastrophic Forgetting in Language Transfer via Model Merging
by Anton Alexandrov, Veselin Raychev, Mark Niklas Müller, Ce Zhang, Martin Vechev, Kristina Toutanova
First submitted to arxiv on: 11 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This research paper proposes a new method, Branch-and-Merge (BaM), for adapting open-weight large language models (LLMs) to different languages. The goal is to improve the performance of LLMs on various tasks in non-English languages while minimizing the loss of their original capabilities. BaM iteratively merges multiple models fine-tuned on a subset of training data, resulting in lower magnitude but higher quality weight changes that reduce forgetting and maintain learning. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper shows how to improve language models for different languages without losing what they can do originally. It uses a new method called Branch-and-Merge (BaM) which combines multiple models trained on parts of the data to get better results. This helps keep the original abilities of the model while still learning about the new language. |
